Type 1 diabetes is an autoimmune disease characterized by T cell-mediated destruction of the pancreatic islet beta cells and the subsequent inability of the patient to produce the insulin required to properly regulate glucose metabolism. CD8 T cells are important contributors to the demise of beta cells in both the non-obese diabetic (NOD) mouse model of the disease and type 1 diabetes patients. A more complete understanding of the epitopes recognized by these T cells is needed to facilitate the development of antigen- specific strategies to interfere with their pathogenicity, as well as assays to monitor autoimmune activity in at- risk individuals, islet transplant patients, or those undergoing intervention protocols. Our proposed project, though self-contained and modest in scope, addresses these unmet needs. In Aim 1, we will define the HLA- A2-restricted T cell response to the important autoantigen islet-specific glucose-6-phosphatase catalytic subunit-related protein (IGRP) using an innovative HLA-A*0201-transgenic NOD mouse model that expresses human IGRP in the absence of its murine counterpart and also lacks murine class I MHC molecules, so all CD8 T cells will be restricted to HLA-A2. We will use the islet-infiltrating CD8 T cells from these newly developed, diabetes-susceptible mice to screen a library consisting of all possible 8-, 9-, 10-, and 11-mer peptides that can be derived from human IGRP. This unbiased strategy will allow us to obtain a comprehensive picture of the spontaneously arising HLA-A2-restricted T cell response to this autoantigen. In Aim 2, we will characterize these human IGRP epitopes in terms of their MHC binding properties and the pathogenicity of their cognate T cells. Candidate CD8 T cell epitopes from human autoantigens have often been chosen for testing of patient reactivity based on their predicted ability to bind an MHC molecule of interest. However, an emerging view is that the peptides recognized by autoreactive T cells often bind poorly to MHC, allowing escape from immune tolerance mechanisms. In Aim 1, we will be identifying epitopes of IGRP in a non-biased manner. Thus, we will have the opportunity to examine the HLA-A2-binding ability of the identified epitopes and test the hypothesis that low-affinity binding will be favored. To extend the potential usefulness of our epitopes to patient that are HLA-A2-negative, we will also determine their ability to bind related MHC molecules. Finally, to investigate the pathogenicity of the T cell specificities identified in Aim 1, adoptive transfer of peptide-specific CD8 T cells will be employed. Characterization of the human IGRP epitopes in these ways will allow the identification of those most likely to be useful as therapeutic targets and markers of pathogenic autoimmunity.